Another paper finds that Google trends data are useful for improving nowcasts (meaning forecasts of current events for which data has not been released), this time of US consumption data:

“In this paper, we investigate whether the Google search activity can help in nowcasting the year-on-year growth rates of monthly US private consumption using a real-time data set. The Google-based forecasts are compared to those based on a benchmark AR(1) model and the models including the consumer surveys and financial indicators. According to the Diebold-Mariano test of equal predictive ability, the null hypothesis can be rejected suggesting that Google-based forecasts are significantly more accurate than those of the benchmark model.”

Like other papers nowcasting with Google data, the models used are simple. In fact, when the authors use a model with more explanatory variables, the Google data no longer improves predicability.

Improvements on simple AR(1) models are interesting, but I am very curious to know whether Google data is useful in improving the large, complex, and finely tuned forecasting models that professional forecasters use. An and Google partnered study on this issue seems like an obvious win-win. What’s the hold up?